E-Racing Together How Starbucks Reshaped and Deflected Racial

Public Relations Review 45 (2019) 101773

Contents lists available at ScienceDirect

Public Relations Review

journal homepage: www.elsevier.com/locate/pubrev

E-Racing together: How starbucks reshaped and deflected racial conversations on social media T ⁎ Alison N. Novaka, , Julia C. Richmondb a Rowan University, 301 High Street, 325, Glassboro, NJ, United States b Drexel University, 3141 Chestnut Street, Philadelphia, PA, United States

ARTICLE INFO ABSTRACT

Keywords: In March 2015, Starbucks introduced its #RaceTogether campaign to encourage patrons to Starbucks discuss race and ethnicity in global culture. Public reaction to #RaceTogether was largely critical Twitter and resulted in Starbucks' abandoning the campaign within a year. Through an analysis of 5000 Reputation management #RaceTogether tweets, this study examines how users engaged the campaign and each other. Race This study draws three conclusions. First, most #RaceTogether posts featured extremist and Brand racist positions. Second, #RaceTogether posts deflected race conversations and critiqued the Social media organizations role in national racial discourses. Finally, posts in the digital space critiqued Starbucks as a location for inter-racial dialogue because of brand perception. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

1. Introduction

In March 2015, Starbucks introduced its #RaceTogether campaign designed to encourage patrons to openly discuss and debate the contemporary treatment and place of race and ethnicity in global culture. In the aftermath of police violence targeted towards American minorities (and specifically Black Americans), the corporation intended to provide a physical and digital space forcus- tomers to reflect on recent events (Hernandez, 2015). While conceivably coming from a good place, the public’s reaction to #Ra- ceTogether was far from favorable and resulted in Starbucks’ abandoning the campaign within a year. Media criticized Starbucks’ efforts and questioned if the campaign produced more tensions between racial groups instead of helping heal national divisions (LaMonica, 2015). Ultimately, Starbucks experienced a crisis just before the 2015 winter holiday season due to digital protests and public backlash (Peterson, 2015b). This study looks at how the public digitally engaged with the #RaceTogether campaign and critiqued its ability to enable discourse on race. While previous research examines the campaign from a critical race perspective, this study uses Twitter data to investigate public response and adoption of the campaign (Logan, 2016). Its findings hold implications for those studying interracial communication, digital public relations campaigns, and new media. An immediate challenge to the #RaceTogether campaign was the public perception of the motivations behind the corporatization of the facilitation of a national dialogue on race. In large part, the public backlash was a result of the perception of a disingenuous strategy from a racially uniform, wealthy, and affluent retailer. As a result, the failure of the #RaceTogether campaign is critically important to scholars who examine digital discourses about race. Despite the public dislike for corporate messaging, previous work suggests that the campaign did bring a diverse group of people together and into agreement (if only for their criticism of Starbucks)

⁎ Corresponding author. E-mail address: [email protected] (A.N. Novak). https://doi.org/10.1016/j.pubrev.2019.04.006 Received 6 August 2018; Received in revised form 7 February 2019; Accepted 16 April 2019 Available online 28 April 2019 0363-8111/ © 2019 Elsevier Inc. All rights reserved. A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773

(Peterson, 2015b). Thus, looking at public engagement with the campaign can reveal the mechanisms for this inter-racial conversation oriented towards the corporatization, or the integration of for-profit organizations into previously public-domain cultural goods and concepts, of a race-based dialogue (Hayhurst & Szto, 2016). This article uses a discursive analysis approach to study public engagement on social media. Through an analysis of a collection of 5000 public tweets including #RaceTogether, this study examines how users engaged with the Starbucks campaign and each other. Posts were collected from March 15, 2015 to November 1, 2015 (the dates of the active campaign, before its end). These tweets reveal how the public used the campaign to adopt and shape race-based discourses, as well as the corporatization of racialized issues. This study draws three conclusions. First, most #RaceTogether posts featured extremist and racist positions on race. Second, #RaceTogether deflected conversations on race and instead produced a national dialogue on who (or what) should encouragena- tional racial discourses. Finally, although coffee shops historically serve as a meeting place for civic conversations for centuries, posts in the digital space critiqued Starbucks as a location for inter-racial dialogue because of brand perception. These three findings reinforce the #RaceTogether campaign's ability to both build community and facilitate national conversations on race.

1.1. Public engagement and digital communication on race

Scholars note that digital media provides a space for users to engage each other on difficult topics such as race, gender, and age- based divisions in society (Novak & El-Burki, 2016). Although there are many reasons cited for the digital shift in the public sphere, users cite their own dissatisfaction with traditional or physical routes to civic engagement (Novak, Johnson, & Pontes, 2016). The digital space offers an alternative structure for citizens to engage each other in difficult topics that center on identity,policy,or justice. This digital shift produces digital dialogic communication, or a typology that allows organizations to engage members of the public directly and solicit feedback for mutual adjustment (Taylor & Kent, 2014). Dhanesh (2017) outlines the various approaches of this scholarship, particularly emphasizing the need for practitioners to recognize and utilize engagement as a tool for relationship management and improvement. This emphasis is drawn from engagement’s definition, which articulates “an understanding of dialogue with users” (Dhanesh, 2017, p. 926). This includes how top management communicates on social media with members of the public to explain corporate decisions and solicit feedback (Men, Tsai, Chen, & Ji, 2018). However, online communication extends beyond organizational feedback, it is also used by organizations to cultivate digital communities and form collective action on social issues (Watkins, 2017). Despite the plethora of studies examining race in a digital context, Daniels (2013) confirms that scholarship has no conclusions regarding if social media succeeds in equalizing races (even when that is a discursive expressed intent). Graetz, Gordon, Fung, Hamity, and Reed (2016) argue that digital spaces further entrench racial differences and divide races from each other: digital conversations on race statistically focus on division, rather than proactive ways to repair or work towards racial equality. Campos- Castillo (2015) argues that as early as 2007, social media spaces encouraged users to see themselves different than others, particularly based on race. Early social media sites encouraged profiles where users identified themselves by gender, age, and race/ethnicity. These sites segmented identity into these few categories, thus confirming that those were the most important factors that divided society (Campos-Castillo, 2015). Tynes and Mitchell (2014) argue that Twitter serves as a space for African American youth to discuss their treatment by mainstream culture (such as pop culture representations). Although conversations like these may help individuals feel connected to larger communities, they do little to challenge or address the issues being discussed (Jackson, von Eye, Fitzgerald, Zhao, & Witt, 2010; Tynes & Mitchell, 2014). Other scholarship argues that movements proliferating from digital spaces, such as “Black Lives Matter” and the “99%,” demonstrate the ability of digital conversations to organize and make meaningful contributions to culture and society (Carney, 2016; Hooker, 2016; Larson, 2016; Rickford, 2016). In this vein of scholarship, digital conversations on social media platforms like Twitter, result in the physical organization and mobilization of like-minded communities who seek policy and social change (Larson, 2016). The digital space is a type of jumping-off platform for membership to negotiate goals, identity, and actions of a movement across physical boundaries (Carney, 2016). Novak et al. (2016) found that Twitter functions as a space for users who feel left out or rejected by traditional routes to political and civic engagement. The network provides a space for individuals who were previously silenced or rejected by the traditional political parties, candidates, or organizations. The relative newness of social media platforms serves as a challenge for scholars who seek to examine the impact or effect of digital race-based conversations. Sharma’s (2013) landmark study on Black Twitter argues that it could be decades before we can truly test the impact of race-based conversations because of the slow-moving pace of social change. Porter, Anderson, and Nhotsavang (2015) note that studies of Twitter conversational engagement and its effect on mass society need time to unpack the complexities of how social changes such as racial equality take place. Thus, drawing a consensus or rectifying the two veins of scholarship described above will take time (Porter et al., 2015; Sharma, 2013). The growth in accounts and users who identify with the Black Twitter community demonstrates how a campaign like #RaceTogether intended to draw on existing networks of users for increased attention. Freelon, McIlwain, and Clark (2018) argue that existing digital communities mobilize faster in response to controversies, rather than individual users- thus making these groups appealing audiences for corporate campaigns. Gallagher, Reagan, Danforth, and Dodds (2018) notes this is likely because of the protest-based origins of Black Twitter, as users responded to injustices in policing. Of importance: the ties between the growth within the Black Twitter digital community and the pattern of using current events or trending topics to focus attention to issues of social injustice (Ray, Brown, & Laybourn, 2017; Williams, 2015). While examining how the campaign was specifically used within this community is outside the scope of this study, Black Twitter may be partially responsible for the salience of the #RaceTogether campaign (Logan, 2016). Significantly, no studies have specifically examined the relationship of #RaceTogether and Black Twitter,

2 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773 although this study identifies several areas where the community and campaign converged. Part of the complexity with this research is understanding the context of how users tweet, particularly when discussing race. Marwick and boyd (2011) note that when users tweet, they do so for an imagined audience, which impacts how they construct messages. For example, if users are writing for other-like minded individuals or within a community, they will use internally re- cognized jargon and slang, reference collective memory or experience, and invoke community rules and norms (Marwick & boyd, 2011). However, when writing for an external or hostile audience, users write defensively, providing evidence such as links for more information, and sometimes in an attacking-manner (Gruzd, Wellman, & Takhteyev, 2011). The difference in this messaging and style of communication may account for some of the differences in the perception of the effectiveness of digital race-based conversations. The #RaceTogether campaign reflects an opportunity to study how users construct tweets for internal or external imagined audiences, and how this may impact the direction and impact of a strategic campaign. Uses and gratifications research on social media similarly hold implications for understanding how customers mayengagea corporate campaign on controversial issues. Whiting and Williams (2013) found 64% of Twitter users adopt the platform primarily for entertainment purposes, including humor. Further, Neuendorf, Skalski, Jeffres, and Atkin (2014) argued that users were quick to use humor and irony within tweets on culturally difficult or controversial topics. In fact, the researchers suggested that corporate organizations adopt humor in social media campaigns to align with this uses and gratifications finding (despite earlier findings that many corporations failed to use the humorous potential of digital communication) (Neuendorf et al., 2014). Research from Fraustino and Ma (2015) may provide Starbucks with a way to anticipate the public reaction to #RaceTogether. Their work on the Center for Disease Control’s Zombie Apocalypse campaign suggests that corporations who attempt to insert themselves into sensitive cultural conversations should anticipate a hostile, humorous, or ironic reception in a digital space. Despite these findings, Fraustino and Ma (2015) call for more scholarship of this hostile reaction so that campaign development can be better informed.

1.2. Organizations and race initiatives

Despite the lack of consensus regarding the effect of race-based social media conversations, several organizations have turnedto the digital space to facilitate and engage customers. Although Starbucks’ #RaceTogether campaign was probably one of the most popular examples of this digitization of race-based conversations, it was far from the first organization to experiment with the digital format. A case from just two months before in 2015 likely made Starbucks’s Twitter campaign more salient and newsworthy, foreshadowing the reception of #RaceTogether. The New York Police Department encouraged residents to share personal stories on Twitter regarding their relationship with the police (Hayes, 2017). #MyNYPD went viral, as users shared hundreds of videos, images, and stories of alleged unethical police conduct, primarily stemming from racial tensions and conflicts (Hayes, 2017). In the week following, the New York Police Department apologized for the insensitive campaign and encouraged the public to engage in a thoughtful and meaningful conversation on race and police work in the future (Jackson & Foucault Welles, 2015). The #MyNYPD controversy predated the #RaceTogether campaign, and foreshadowed upcoming Starbucks tensions. Users readily criticized the New York Police Department for its inability to recognize its own part in current racial tensions and inequality (Jackson & Foucault Welles, 2015). Even after its apology, the organization was criticized as disingenuous because of the failure to make changes to policies and behaviors that may unfairly target minorities (Dean, 2014). Considering recent police brutality scandals in other major cities around the United States, #MyNYPD was viewed as insensitive and emblematic of an organization who was unable to consider its own role in controversies (Dean, 2014). Similar criticisms would be made of Starbucks as it launched #Ra- ceTogether. Despite these similarities, there are also notable differences between the cases. First, the NYPD is not a for-profit organization, meaning its communication tactics do not reflect a bottom-line orientation, unlike a major corporation like Starbucks. In addition, unlike the #RaceTogether campaign, #MyNYPD did not explicitly aim to create discourses on race, it was an unintended outcome. Instead, #MyNYPD focused on improving public engagement with the police force. As Starbucks expressed in its own published communication on the topic, the organization aimed to improve the amount and quality of racial discourses, not primarily its public image- ironically, this corporate insertion into the social issue was a central vein of public criticism (Starbucks, 2015). Scholars note that users are quick to reject corporate campaigns that seem to take advantage or profit from social conflicts. Vos and Li (2013) argue that social media platforms emerged as spaces for consumers to vent frustrations over faulty or problematic campaigns. Customers react to crass commercialization, where a corporation is seen as adopting a trend to make a profit, rather than meaningfully engage with the community or culture (Vos & Li, 2013). Users particularly reject these campaigns and turn to social media to share their frustrations with others. Dorfman, Cheyne, Friedman, Wadud, and Gottlieb (2012) argues that these campaigns are critiqued for lack of authenticity. For example, consumers rejected companies that sprang up after the September 11 terror attacks, profiting from apparel with “Never Forget” on them(Dorfman et al., 2012; Potts, 2012). Organizations that use crass commercialization are considered inauthentic and lacking self-awareness, thus making them easy for consumers to reject in a digital space (Dorfman et al., 2012; Potts, 2012). Potts (2012) reflects that crass commercialization does not mean the company will failor stop. In many cases, digital outrage does not hurt sales, and may be considered method that improves recognition and salience (Potts, 2012). Potts (2012) notes some organizations even adopt these practices because online users will draw attention. Research on race in strategic campaigns demonstrates the difficulty of organizations developing campaigns without reflecting on their own diversity, equality, and position in racialized society (Heath & Waymer, 2014; Waymer, 2012). Waymer’s (2012) volume identifies campaigns that struggled to recognize responsibilities in race and class-based exclusionary practices. For example,during the 2011 Occupy movement, social media activists asked Occupiers to identify personal moments of inequality and injustice on social media (Waymer, 2012). In return, other users adopted the Occupy hashtag to identify how protesters were using historical Native American territory to protest issues of social justice, without reflecting on their own history of oppressive behavior (Waymer, 2012).

3 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773

Again, although not a corporation in the same sense as Starbucks, the outcry from Native American advocates towards the irony of the Occupy protest occurring on former territory showcased the problems organizations and leaders face when they ignore their own histories of racial inequality. Audiences and members of the public often reject campaigns or messaging that fail to adequately articulate and rectify their participation in race-based inequalities. Winter (2014) work further calls for organizations to work closely with public relations practitioners to understand perceptions before developing and implementing campaigns that insert organizations into race-based dialogues or current events. Despite this call to action, Winter (2014) reflects that many organizations (even ones with the bestof intentions) fail to do this, thus producing more reputational harm than social good. #RaceTogether’s background sheds some light into how the campaign was designed and its intentions in contemporary race-relations.

1.3. #RaceTogether

In Spring 2015, Starbucks launched #RaceTogether to initiate conversations about race in the United States and around the world. This campaign, largely pushed by CEO Howard Schultz, was intended to follow the lead of the #BlackLivesMatter movement in reacting to a recent grand jury decision in Ferguson, Missouri (Peterson, 2015c). The decision not to charge Darren Wilson, the police officer who killed 18-year old Michael Brown, sparked national conversations and protests surrounding issues ofraceand culture. However, unlike other digital responses such as the #BlackLivesMatter movement, #RaceTogether would be propagated by a corporate entity with an implicit objective to sell products. In its beginning, #RaceTogether was debated even within the company. During the planning phase of #RaceTogether, Starbucks executives were afraid of the potential backlash (Shah, 2015). The main concern was with the company’s own multicultural shortcomings (Shah, 2015). It was Schultz who supported the idea and saw it through to realization, despite many internal objections (Hernandez, 2015; Shah, 2015). Cory duBrowa, Senior Vice President of Communication was one of the few individuals who initially supported the campaign, although after he left the company two years later, he remarked that he felt pressured to support it by Schultz (Kanski, 2016). It is important to note that this is contradictory to the official Starbucks announcement, which suggests that the campaign originated from baristas who wanted to help people engage in critical thinking on race (Starbucks, 2015). However, later duBrowa, suggested that any internal or employee support of the campaign was probably pressured by Shultz. (Peterson, 2015b). Taranto (2015) noted that Starbucks public relations team either failed to serve as ethical watchdogs in the days before #RaceTogether’s implementation, or were totally ignored by Schultz. Peterson (2015c) notes that the immense secrecy behind #RaceTogether was drastically different than the traditional transparency of the organization, suggesting the campaign hadlittle employee support from the start. Based on this insight, it is likely that Schultz was acting as a transformational leader, one who holds a specific vision and set of goals for the organization, rather than one who performs inclusive leadership(Werder & Holtzhausen, 2009). Because Transformational leaders “appeal to followers’ ideals and moral values,” they also take risks and are likely to create campaigns that aim for a larger-good, including social or cultural good (Werder & Holtzhausen, 2009, p. 406). However, these campaigns require strong communication and often risk misinterpretation unless the organization exhibits “openness, dialogue, frankness, careful listening, and informality” (Werder & Holtzhausen, 2009, p. 406). As described in the next section, Starbucks failure to demonstrate each practice was an insurmountable problem for the campaign. On March 15, Starbucks started the campaign by requiring employees to write #RaceTogether on the cups of customers to encourage race conversations in the café and online (Hernandez, 2015). #RaceTogether was in-line with the corporate goals and the Starbucks initiative to make its coffee shops a “third place” for people (other than home and work)(Starbucks Company Information, 2019). The ideal third place would foster a sense of community and civic engagement beyond work and home life. With this goal in mind, Starbucks anticipated it could use its 22,000 domestic stores as a springboard for further national discourse about issues of race in the United States. In this respect, the initiative was successful in an unexpected way; Starbucks intended to be the setting for a national conversation, although it also became the subject of it (Hernandez, 2015). Soon after the campaign started there was major backlash from customers. Coffee drinkers used #RaceTogether overwhelmingly to criticize the company (Sanders, 2015). Some tweets highlighted that Starbucks was not diverse enough as a company and should look inward when contemplating issues of race. Others found that Starbucks put employees in an inappropriate position, which they should not be required to handle. Customers were also put off by the simplistic treatment of conversation about racial inequality. Aside from customers, public figures added to the backlash. Gwen Ifill, a PBS anchor, poked fun at Starbucks by tweeting “Honest to God, if you start to engage me in a race conversation before I’ve had my morning coffee, it will not end well (Nocera, 2015).” Further, Harvard University professor Dina Pomeranz pointed out in a tweet “Starbucks' #RaceTogether invites customers to talk about race. Uses only white hands in related photos.” Late night comedians also reacted to the #RaceTogether campaign (Peterson, 2015c). Cast members of Saturday Night Live (SNL) performed a parody commercial which mockingly proposed a #Genderflect campaign where gender and sexuality would be discussed at Pepboys. Following a clip of Schultz introducing #RaceTogether, Last Week Tonight with John Oliver claimed that “it is pretty clear, no one has said no to this guy in 25 years.” The attention did not go unnoticed by Starbucks officials. During the backlash, Cory duBrowa, Senior Vice President of Communication at Starbucks deleted his Twitter account due to his receipt of thousands of attacking personal messages (Peterson, 2015a). By March 30, Starbucks discontinued #RaceTogether in stores, adding that they looked forward to continuing the hashtag on Twitter until November (but would no longer write #RaceTogether on cups) (Peterson, 2015c). On November 1, 2015, Starbucks ended the campaign, citing the negative public reaction (Peterson, 2015c).

4 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773

1.4. Campaign messaging

As brands prepare social media campaigns, they must carefully consider the messaging strategy best positioned to garner public engagement. Hazelton (1992) identifies six message strategies traditionally utilized by practitioners: informative, persuasive, facil- itative, coercive, co-operative problem solving, and bargaining. In the case of #RaceTogether, the campaign aimed to facilitate conversations between customers and members of the public on the topic of race. Previous work identifies that facilitation message framing helps audience members collaborate on socially difficult topics, such as teen pregnancy, health issues, and environmental advocacy (Devine, Bull, Dreisbach, & Shlay, 2014; Guo & Saxton, 2014; Vidrine et al., 2012). Starbucks even used “facilitate” in its early tweets instructing users on how to use #RaceTogether: “We aim to facilitate a conversation.” (Peterson, 2015b). This is ap- parent in other forms of official communication from Starbucks during the campaign. In a digital announcement explaining the campaign, Schultz reflected, “it is an opportunity to begin to re-examine how we can create a more empathetic and inclusive society– one conversation at a time” (Starbucks, 2015). Important to the facilitation strategy, is the required labor associated with cultivating dialogue and discussion. Through this strategy, organizations must go beyond providing a topic for conversation, but also curate and continually engage participants. In Starbucks case, writing #RaceTogether is not enough to perform a facilitation strategy; for a fully effective campaign, the organization must also participate in the conversation. Liu, Lai, and Xu (2018) warn this is the most difficult part of the facilitation strategy: the continual effort required throughout the campaign, not just the launch or distribution of themessage. Thompson and Cupples (2008) argue that the risk of a facilitation message strategy is that an organization may not be prepared to perform the labor of internal reflection and full participation in the proposed dialogue, leading audience members to question the authenticity and motivation behind the campaign. As noted earlier, Fraustino and Ma (2015) suggest organizations that insert themselves into sensitive cultural issues or events can expect a negative public reaction unless the message strategy is fully developed and the organization is willing to perform the ongoing labor of facilitation. Public reaction to #RaceTogether and Starbucks serves as an ideal case study to examine how engagement with a race-based corporate campaign manifests in an online space.

2. Methods

This study uses a discursive analysis approach to examine how the #RaceTogether campaign impacted digital conversations on race and ethnicity. Previous research suggests discursive analysis is a strong methodological tool for studying Twitter because of its ability to examine a set of tweets holistically, rather than in smaller isolated parts (Mengibar, 2015). Tweets are emblematic of larger dialogue; thus they need to be read and interrogated for patterns, terms of reference, and the language of significance in order to understand how a digital community converses on a topic. Gee’s (2011) seven meaning making tasks are used to discursively examine #RaceTogether tweets. As used in other Twitter studies, Gee’s (2011) approach supports the aims of this project, looking for how a concept is interrogated by users within a limited technological space (in this case, 140 characters or less). The project adopted the procedures from previous work using Gee (2011) to identify discourses on Twitter (Altoaimy, 2018; Paskevicius, Veletsianos, & Kimmons, 2018; Dimitrakopoulou & Boukala, 2018). In these studies, Gee’s (2011) framework is applied by researchers as they independently read the data. Each of seven meaning making tasks and corresponding question (see Appendix A) are applied by looking for examples of tweets that demonstrate how a user understands, interprets, or communicates on a specific topic and answers the question. For example, in Altoaimy’s (2018) work on user reactions to the lifting of a driving ban on women in Saudi Arabia, the researcher read tweets looking for examples of how male, female, and governmental identities were constructed. Tweets were segmented into categories that reflected each of the seven practices and revealed larger patterns in the identification and value of drivers (Altoaimy, 2018). Because of a shared set of defi- nitions of meaning making tasks, research using Gee’s (2011) framework allows scholars to study a digital dataset on any topic to identify discursive practices and construction. Researchers independently read a random sample of 5000 public tweets from March 15, 2015 to November 1, 2015 (the dates of the active campaign, before its end) and applied Gee’s (2011) seven meaning making tasks: patterns of significance, practices, identities, relationships, politics, connections, and sign systems/knowledge (see Appendix A for description and examples of each practice). After reading the dataset, the researchers then discussed and developed three discourses (See Appendix B for step by step procedure). For qualitative reliability, quotes are included to support each finding.

3. Findings

3.1. Discourse one: extremist and racist positions

As noted by Logan (2016), part of the controversy surrounding #RaceTogether was the extremist and racist digital conversation that appeared throughout Twitter. Johnston (2015) notes that scholars can identify extreme or racist online content by examining the context of use and prolonged exchanges and interactions. Rather than use the hashtag for open and affirming race-based conversations, trolls took to Twitter to engage and upset users. Journalists noted these extreme tweets as an unforeseen consequence of inviting the public to engage a topic so closely related to controversial and consequential current events (Taranto, 2015). Users who adopted the #RaceTogether campaign and referenced ongoing police brutality or the Black Lives Matter movement were the targets of digital trolling. For example, this exchange featured: “User 1: Not sure @Starbucks gets how they make racial tensions and propels the needs for BLM, #RaceTogether won’t fix it. User 2: You f***ing idot, Black Lives Matter wont matter in the future. User1:excuse

5 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773 me, I don’t remember asking you. Show some respect for BLM. Use 1: too scared to say anything back? (Isa, 2015).1 What makes these incidents of trolling versus regular exchanges is the lack of long-term engagement between users (O’Callaghan, Greene, Conway, Carthy, & Cunningham, 2015). For the trolls, quickly engaging the user, angering or upsetting them, and then backing-off is a practice of digital communication. Rather than respond to user 1 again, user 2 (the troll) backs-off, knowingtheygot the attention of user 1 and potentially upset them, which is the goal of the conversation. The goal is not an authentic or thoughtful conversation on Black Lives Matter and race. Trolling behaviors are also long-term practices within the #RaceTogether dataset. User 2 (the troll) similarly engaged other users over a two-month period using the hashtag before deleting his/her account- conceivably to re-create it under a new alias. Similarly, other trolls adopted this behavior, briefly engaging users with inflammatory comments and then backing-off to watch theeffectof their remarks. Other users even reflected on how the #RaceTogether campaign inspired them to challenge and “troll” Starbucks employeesin physical locations. One user wrote “My favourite Starbucks troll is to ask the cashier if they want to talk about race while thy warm up my banana bread #RaceTogether” (@Squeekybb, 2015). Although it is unclear if this user followed-through or implemented this behavior, many tweets demonstrated how customers planned to physically object or critique employees who wrote #RaceTogether on their cups. Beyond trolling, there were also many comments that questioned the legitimacy of contemporary concerns on race and the need for such a campaign. Often, these extremist positions could be classified as “racist” or adopting negative stereotypes of minorities. For example, one user wrote: “if they [African Americans- derived from an earlier tweet] spent half as much time working as they did complaining on Twitter = no #RaceTogether” (Some Place Simple, 2015). Again, by adopting negative and false stereotypes, comments like this one critique both the #RaceTogether campaign and the legitimate concerns of African Americans regarding safety and equality. Tweets like this aim to delegitimize the purpose of #RaceTogether as well as other contemporary social movements that aim to encourage equality and racial justice. However, they also seek to shut down race-based conversations by deflecting purpose and redirecting attention. This similarly appeared in discourse two, as users re-directed the hashtag’s purpose to critique and converse on Starbucks, rather than race. Although the majority of tweets were negative in tone (for the variety of reasons above), there are some instances where the public supported Starbucks’ campaign and tried to use it to cultivate race-based discussion not centered on corporate criticism. For example, one user posted, “It’s about time that we really talk about underlying issues in current events. Good for @Starbucks #RaceTogether for helping motivate us” (Bangz, 2015). This message of support mirrored similar ones that thanked Starbucks attention to the issue. However, instances of appreciation for the corporation were rare, and there were no instances where moments of thanks were coupled with personal reflections on race. Other users adopted the hashtag to share their own perspectives and experiences with race: “I’ve been thinking a lot about how police against MB’s skin color was unfairly justified for his death, we need to love each other even if we dif #RaceTogether” (Burge, 2015) Here, many users provided specific examples, like the death of Michael Brown to reflect on the role of race in society. Again, instances of genuine reflection were rare and often produced noactual responses (in the form of replies, retweets etc.). In addition, there were individuals who used the hashtag in order to find out more information. For example, “Why is everyone trying to #RaceTogether?” (Hauer, 2015). There were many instances of people who wanted to know why the hashtag was trending on Twitter. Others sought more information on its controversial elements. For example, “I think its good @Starbucks wants us to talk about #RaceTogether. Why is everyone upset that coffee is motivating us?” (Whitt, 2015). Again, there were few replies to user asking for more information, which may be because of the popularity of criticizing Starbucks rather than delving into race-related issues through discussion and debate.

3.2. Discourse two: deflecting race conversations

Throughout Twitter, there was a generalized consensus that Starbucks’ campaign was not a desired space for critical conversations on race. Rather than discuss race or ethnicity and its place in current global culture, tweets focused on how much users did not want to engage the topic or openly discuss race using the hashtag. One user summarized this desire to avoid the purpose of #RaceTogether with tweets listing her priorities: “What I want to do: 1. Eat Pizza 2. Pay off my loan debt… 77. Hang out with Rihanna… 895.Talk about race at Starbucks #RaceTogether” (Rasheed, 2015). Here, the user deflects the conversation and uses the hashtag to share how much she does not want to engage the topic within the commercialized Starbucks space. Most tweets using the hashtag outwardly rejected the idea of discussing race online, instead turning their attention to discussing Starbucks’ public relations nightmare. One user wrote, “Maybe #Starbucks actually wanted to get people of all races & ethnicities to join hands and make fun of #RaceTogether. Well played” (Due, 2015). Tweets like this one, although using the hashtag, failed to fulfil the purpose of the campaign. In fact, most of the tweets using the subject instead reflected on the brand, notrace/ethnicity. From a public relations perspective, this is interesting considering the prolific use of the hashtag, but failure of users to adopt it forits expressed purpose. In this sense, users co-opted the hashtag and instead used it as a mechanism for criticism of the brand. This will be discussed in the next discourse. Users also adopted the hashtag to identify other faults of the Starbucks brand. For example, “Before they write #RaceTogether on cups, can Starbucks just spell my name correctly?" (Minhaj, 2015). Again, this is a type of objection to the initial purpose of the

1 All tweets are copied directly from Twitter and have not been edited for grammar or spelling.

6 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773 hashtag. Rather than a conversation on race, this is a discussion of the brands ability to properly identify and spell customer names. Interestingly, another hashtag developed and was popularized by users alongside #RaceTogether. Although it appeared more organically, through user comments, #NewStarbucksDrink frequently included themes from the #RaceTogether campaign. For example, one user proposed drinks that adopted current and contemporary pop-culture references: “"Mocha Money, Mocha Problems" "No Chai Left Behind" "Black Coffees Matter" and "BY ANY BEANS NECESSARY" (Tracy the Clayton, 2015). Although, this tweet contains references to the Black Lives Matter movement, it is far from a nuanced and direct conversation on race that #RaceTogether hoped to achieve. Discursively, humor was used throughout tweets to re-direct conversation from difficult racial topics. As noted by Whiting and Williams (2013), humor remained a dominant response strategy for users attempting to respond to the campaign. Although pivoting their attention to critiquing Starbucks, humor was used as a type of practice to alleviate tension and the difficulty of discussing race. Throughout the dataset, irony and humor were predominantly the way users reacted to #RaceTogether. One user summarized the practice with “As a stand up comedian, #RaceTogether is giving me the best material of my life. Thnx @Starbucks” (Slush, 2015). This approach and practice of humor and irony is discussed in the reflection.

3.3. Discourse three: corporatizing race conversations

For many users, the campaign was a crass and insulting attempt to inject a corporation within a needed national dialogue about race. Users were quick to reject Starbucks as a facilitator of race-based conversation, citing the company’s brand positioning, its history with gentrification, and its problematic quasi-fast food status. Many users adopted an ironic approach to share theirdistaste for Starbucks’ presence in the national dialogue. In some cases, images and memes were uploaded to support user arguments. For example, one user created a meme of a group of white Starbucks employees with “Welcome to Starbucks, Let’s discuss your white privilege” written on it (BBC Trending, 2015). Meme’s like this one, where Starbucks is characterized as predominantly white, middle class, or privileged were common. Through irony, users critiqued the position of Starbucks and its racial association with the white middle and upper-classes. This discourse was supported through comparisons of Starbucks to other cultural manifestations of race and identity. One user wrote “#RaceTogether is what happens when a 1%-er without any actual anti-racist education or training has a mid-life ‘white man's burden’ crisis” (Flemming, 2015). Tweets often associated Starbucks as an oppositional force to the 99% movement and emblematic of the 1% elite. Users expressed their discomfort with a brand that is marketed as exclusionary and elitist, trying to facilitate a conversation about exclusionary and elitist cultural divisions. This was supported through tweets such as: "’Barista: Your total is $5.45 Me: You can just put that on my reparations tab. Thanks. #raceTogether’" (Stafford, 2015) or “Being a barista is hard enough. Having to talk #RaceTogether with a woman in Lululemon pants while pouring pumpkin spice is just cruel” (Oluo, 2015). Tweets like these characterize Starbucks as part of the larger racial divisions that stem from historical inequalities and manifest within contemporary brands. For these users, Starbucks is part of the problem, thus making its #RaceTogether campaign insulting and even entrenching racial divisions. Other tweets similarly referenced the Black Live Matter movement as well as Black Twitter. But rather than focus on race as a cultural issue, these tweets focused on Starbucks as an agent of change. For example, “Does Starbucks think that it can do better than what all of #BlackTwitter does with #RaceTogether?” (Ensor, 2015). Tweets like this demonstrate a type of territory marking, where users were quick to emphasize Black Twitter’s legitimacy as a motivator of digital communication and arbiter of discussing race on social media. Importantly, tweets that also used #BlackTwitter focus primarily on criticizing Starbucks motivation for the campaign, often calling the company “profitiers” or “crass” or “uneducated” on issues of race. These terms of reference and identity discourses demonstrate the frustrations of users who viewed Starbucks as using race for profit rather than authentically interested in dialogue. As noted by Freelon et al. (2018), tweets using Black Twitter similarly critiqued organizations like McDonalds and Walmart for what was considered inauthentic support of the Black Lives Matter movement performed for publicity rather than social justice. Other users criticized the ability of any brand, particularly one that emphasizes speed, to facilitate a meaningful conversation on race. One user tweeted: “Not sure what @Starbucks was thinking. I don’t have time to explain 400 years of oppression to you & still make my train. #RaceTogether” (April, 2015). Here, users suggested the campaign was uninformed about the social and cultural history of racism in America. The brand, and its position as a quasi-fast food chain, limits the ability of customers and its employees to have meaningful and thorough conversations on any topic, let alone race. In addition, the food industry’s supply-chain and products were viewed as problematic for any food-related company hoping to engage racial conversations. For example, “#RaceTogether should matter to @Starbucks : Coffee and sugar were 2 of the 3 drugs (w/tobacco) that started this mess in the firstplace”(Carr, 2015). Comments like these reinforce the lack of cultural sensitivity the campaign portrayed, as it continued to serve products (and profit from products) that historically disadvantaged and enslaved African Americans. For these users, without proper self-reflection on its own role in racial division, Starbucks would never be able to propel or facilitate national conversations on race. Throughout this discourse is the dominant criticism by users that Starbucks was disingenuous when launching the #RaceTogether campaign, as evidenced by a lack of ongoing critical reflection by the company. Rather than an authentic attempt to engagethe public in a much-needed dialogue about the contemporary treatment of race in America, users believed this was a commercial attempt to insert the brand into a trend. In short, #RaceTogether was a way to profit from current racial tensions manifesting within ongoing incidents of racialized police brutality and the growing Black Lives Matter movement. This orientation towards profit and commercialization was criticized for its crass and unethical co-option of digital culture. One user wrote: “The arrival of @Starbucks is typically a key indicator of gentrification in low-income communities, but #racetogether” (Evans, 2015). Starbucks’ past behaviors and history as a force of gentrification against low income and minority communities directly opposes the expressed goalof

7 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773

#RaceTogether. Thus, users were quick to reject the campaign and criticize Starbucks as inauthentic and crass commercialization. Although criticism of Starbucks appears throughout the dataset and entire span of the campaign, users were particularly angry when Starbucks’ Communications Vice-President Corey duBrowa deleted his account after users angrily tweeted their objections to the campaign. Deleting his account seemingly confirmed criticisms that the corporation was only interested in profiting fromra- cialized conversation, not actually participating in difficult cultural dialogues. One user wrote “@coreydu so are you unblocking/ blocking people who are actually engaging you on a conversation abt race? How #racetogether is a stupid idea?" (Krage, 2015). The company’s inability to support and engage users on the topic was characterized as an insult to those users who wanted to use the digital space to discuss race online. The face-saving measures taken by duBrowa were indicative of a larger problem: that the corporation was unable to carry out the labor of facilitation and hold serious and meaningful conversations on race due to their difficulty. To users, this indicated that the corporation was unprepared for the response and reinforced the perception thatthe campaign was not a deep cultural dive into Starbucks’ role in race relations, but instead a surface-level stunt to gain public attention. By deleting his account, duBrowa was accused of not taking racialized divisions seriously and authentically wanting to engage the public on a difficult issue. duBrowa never formally acknowledged the temporary deletion of his account, and never explainedwhy, although many uses extrapolated that it was because of the thousands of negative responses. The impact of these behaviors is addressed in the next section.

4. Discussion

The three discourses identified within the #RaceTogether tweets suggest that the campaign represents an important areaof inquiry into how a digital race-based campaign is shaped through user engagement. Although Starbucks expressly intended the hashtag to facilitate a contemporary (and admittedly) difficult conversation on race, users had other ideas. Whether users expressed extremist positions, re-directed to critique Starbucks, or created their own associated hashtag, Starbucks’ goal of facilitating digital conversations on race was complicatedly actualized. Theoretically, the rejection of #RaceTogether’s purpose demands inquiry into who holds the power over a digital campaign that uses Twitter and a hashtag as its main communication strategy. Despite Starbucks best efforts to re-direct the conversation to race numerous times, the public redirected attention to the organization. Thus, it was the public who controlled the campaign’s direction. This is critical for the facilitation messaging strategy, and supports Werder and Holtzhausen’s (2009) findings that the public demands organizations continually engage the dialogue, rather than shying away from complicated social issues. #RaceTogether highlights two important implications for corporations hoping to address race. First, the organization should be prepared to engage in the topic in a thoughtful, honest, and reflective way. Starbucks inability to recognize its own role inracial tensions, along with the disappearance of key leadership voices part way through the campaign, frustrated uses who sought a meaningful discussion. Organizations should both be prepared for public critique as well as prepare leadership to partake in the conversation to demonstrate a commitment to dialogue. Second, organizations should carefully consider the timing of the start of a race-focused campaign. The salience of the recent #MyNYPD and recent Michael Brown decisions may have seemed like the perfect time to draw from, but in reality made Starbucks seem inauthentic and advantageous of a difficult social situation. In congruence with Thompson and Cupples arguments (2008), the risk in a facilitation strategy lies in the ongoing labor associated with such critical reflection. Brand managers should carefully consider the timing of a race-based campaign and be prepared to continually engagethe public while considering their own place within the issue. Despite the largely negative use of #RaceTogether against the company, it's important to note that Starbucks reported a financial gain for the 2015 year. CNN Money reported that 2015 was Starbucks’ most profitable year ever, and despite criticisms of the brand’s handling of #RaceTogether, sales were stable throughout that quarter (LaMonica, 2015). As in Potts (2012) study of crass commercialization, campaigns that draw criticism for cultural miscommunication may not ultimately impact the organization’s bottom- line; in fact, the added attention may actually improve sales. This too raises questions for future research, such as how race-based conversations impact customer purchases, particularly brands with high levels of consumer loyalty. Although users were clearly upset with Starbucks’ #RaceTogether campaign, their outspoken digital nature did not impact the physical relationship between customers and the brand. Future studies should consider if this is true for other brands or does the high amount of consumer loyalty to Starbucks differentiate this potential effect? These findings also reinforce Whiting and Williams (2013) and Winter’s (2014) calls for the critical involvement of public relations practitioners in all aspects of campaign development and implementation. While it is unclear if this involvement could have prevented the negative public reaction, it could have better informed organizational responses or anticipated public engagement. The user adoption of humor and irony to respond to #RaceTogether is also an area that needs future research. Rhetorically, humor is often used to diffuse difficult or tense situations where individuals fear insulting or damaging another person’s pointof view. As identified by Marwick and boyd (2011), the imagined audience of tweets on a complicated or difficult topic may encourage a humorous approach. More research is needed to determine if this is the reason why users adopted humor and irony throughout their posts or if this practice occurs in other digital conversations on controversial or difficult issues. There were few tweets that specifically reflected on Black Twitter and #RaceTogether, despite some early media coverage that anticipated the group as a generator of the campaign’s impact. While it is clear that Starbucks hoped to draw on the salience of Black Twitter and protests surrounding police brutality, evidence from this study suggests a limited relationship between the two. While drawing conclusions for this missing link is outside the scope of this study, examining the presence and reception of a corporate campaign specifically within this public advocacy group would serve as a lucrative area of future research. An opportunity for such research on humor and irony in digital conversations may surround Starbucks’ latest campaign called

8 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773

#ThanksToYou. The campaign, launched in January 2017 is designed to thank customers for their ongoing support of specialized populations through purchases. In a January 2, 2017 tweet, Starbucks posted “It's not just coffee, it's helping college students, veterans & farmers find a better future. Good things are happening—#ThanksToYou.” Users were quick to respond with theircri- ticisms of the other places Starbucks donates money, including “corrupt politicians” and baristas who spit in coffee. Studying current and future campaigns from Starbucks may be a way to examine how humor and irony proliferate in the digital space and serve as a practice within difficult contemporary conversations. Other for-profit organizations have similarly weighed in on racial issueswithin society, including Nike’s prolific 2018 campaign featuring Colin Kaepernick. As more for-profit entities produce these campaigns, scholars should focus on how audience reception shapes perceptions of the organization. Years later, Starbucks again faced controversies when two black men were arrested for loitering in a Philadelphia cafe. Upon further investigation, the men were waiting for a friend and it was the store manager who overreacted and called the police. News coverage and social media posts again referenced #RaceTogether as a precursor to the brand’s race problems. Although the corporation shut down for a six-hour racial bias training nation-wide, it is clear that public critique of the organization’s handling and conceptualization of race in society remains.

5. Conclusions

As Starbucks continues to brand itself as the third place for customers to discuss social and political matters, it must reflect on corporate limitations in structuring conversations in civic engagement. While the coffee shops can (and historically have) act asa setting for such dialogue, Starbucks cannot force topics of conversations upon customers without running the risk of scrutiny. This is true in both physical and virtual contexts. In the example of #RaceTogether, Starbucks was scrutinized for attempting to control conversations between individuals and employees. Thus, there was an influx of both physical and virtual trolling, which acted to undermine the intentions ofthe #RaceTogether campaign and redirect the conversation to the Starbucks brand. Trolls were very effective in changing the conversation about #RaceTogether because they were both localized in physical stores and globalized through Twitter. The negative posts elicited responses from baristas who also took frustrations out on Twitter and by corporate leaders such as Corey duBrowa who temporarily deleted his Twitter account. The discourses identified through this study contribute to scholarship concerned with social awareness and corporate entities. Further study should investigate discourses emerging through social awareness, political movements, and public policy campaigns to better understand patterns in discourse and unpack the role of commodity, brand, and power in corporate campaigns. Future research should investigate how corporations, if at all, can positively influence consciousness raising initiatives.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.pubrev.2019. 04.006.

References

[@Squeekybb] (2015). My favourite Starbucks troll is to ask the cashier if they want to talk about race while thy warm up my banana bread. March 17, [Tweet]. Retrieved from#RaceTogetherhttps://www.cnn.com/2015/03/18/opinions/obeidallah-starbucks-race/index.html. Altoaimy, L. (2018). Driving change on Twitter: A corpus-assisted discourse analysis of the twitter debates on the Saudi ban on women driving. Social Sciences, 7(5), 81. https://doi.org/10.3390/socsci7050081. April (2015). Not sure what @Starbucks was thinking. I don’t have time to explain 400 years of oppression to you & still make my train. [@ReignofApril], March 18 [Tweet] Retrieved from#RaceTogetherhttps://twitter.com/ReignOfApril. Bangz, K. (2015). It’s about time that we really talk about underlying issues in current events. Good for @Starbucks #RaceTogether for helping motivate us. [@Kokofasho], March 19 [Tweet]. Retrieved fromhttps://twitter.com/kokofasho. BBC Trending (2015). Starbucks #Racetogether campaign mocked online. March 17, Retrieved fromBBC Newshttp://www.bbc.com/news/blogs-trending-31932351. Burge, D. (2015). I’ve been thinking a lot about how police against MB’s skin color was unfairly justified for his death, we need to love each other even if we dif #RaceTogether. [@iowahawkblog] March 20 [Tweet] Retrieved fromhttps://twitter.com/iowahawkblog. Campos-Castillo, C. (2015). Revisiting the first-level digital divide in the united states: Gender and Race/Ethnicity patterns, 2007–2012. Social Science Computer Review, 33(4), 423–439. https://doi.org/10.1177/0894439314547617. Carney, N. (2016). All lives matter, but so does race: Black lives matter and the evolving role of social media. Humanity & Society, 40(2), 180–199. https://doi.org/10. 1177/0160597616643868. Carr, G. (2015). #RaceTogether Should matter to @Starbucks: Coffee and sugar were 2 of the 3 drugs (w/tobacco) that started this mess in the firstplace. [@AfricanaCarr]. March 19 [Tweet] Retrieved fromhttps://twitter.com/AfricanaCarr. Daniels, J. (2013). Race and racism in internet studies: A review and critique. New Media & Society, 15(5), 695–719. https://doi.org/10.1177/1461444812462849. Dean, S. (2014). MyNYPD Twitter campaign backfires in a big way. The Washington Post, Washington DC: WP Company LLC. Devine, S., Bull, S., Dreisbach, S., & Shlay, J. (2014). Enhancing a teen pregnancy prevention program with text messaging: Engaging minority youth to develop TOP® plus text. Journal of Adolescent Health, 54(3), S78–S83. https://doi.org/10.1016/j.jadohealth.2013.12.005. Dhanesh, G. S. (2017). Putting engagement in its PRoper place: State of the field, definition and model of engagement in public relations. Public Relations Review, 43(5), 925–933. Dimitrakopoulou, D., & Boukala, S. (2018). Exploring democracy and violence in Burundi: A multi-methodical analysis of hegemonic discourses on twitter. Media War & Conflict, 11(1), 125–148. https://doi.org/10.1177/1750635217727311. Dorfman, L., Cheyne, A., Friedman, L. C., Wadud, A., & Gottlieb, M. (2012). Soda and tobacco industry corporate social responsibility campaigns: How do they compare? PLoS Medicine, 9(6), e1001241. https://doi.org/10.1371/journal.pmed.1001241. Due, T. (2015). Maybe #Starbucks actually wanted to get people of all races & ethnicities to join hands and make fun of #RaceTogether. Well played. [@TananariveDue] March 17 [Tweet] Retrieved fromhttps://twitter.com/TananariveDue.

9 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773

Ensor, R. (2015). Does Starbucks think that it can do better than what all of #BlackTwitter does with #RaceTogether. [@richardjensor] March 22) [Tweet] Retrieved fromhttps://twitter.com/richardjensor. Evans, D. (2015). The arrival of @Starbucks is typically a key indicator of gentrification in low-income communities, but #racetogether. [@Duanecia] March 20 [Tweet] Retrieved fromhttps://twitter.com/Duanecia. Flemming, C. (2015). #RaceTogether is what happens when a 1%-er without any actual anti-racist education or training has a mid-life ‘white man’s burden’ crisis. [@Alwaystheself] March17 [Tweet] Retrieved fromhttps://twitter.com/alwaystheself. Fraustino, J. D., & Ma, L. (2015). CDC’s use of social media and humor in a risk campaign—“Preparedness 101: Zombie Apocalypse”. Journal of Applied Communication Research, 43(2), 222–241. Freelon, D., McIlwain, C., & Clark, M. (2018). Quantifying the power and consequences of social media protest. New Media & Society, 20(3), 990–1011. https://doi.org/ 10.1177/1461444816676646. Gallagher, R., Reagan, A., Danforth, C., & Dodds, P. (2018). Divergent discourse between protests and counter-protests: #BlackLivesMatter and #AllLivesMatter. PloS One, 13(4), 90–114. Gee, J. P. (2011). An introduction to discourse analysis: Theory and method (3rd ed.). New York: NY, Routledge. Graetz, I., Gordon, N., Fung, V., Hamity, C., & Reed, M. E. (2016). The digital divide and patient portals: Internet access explained differences in patient portal use for secure messaging by age, race, and income. Medical Care, 54(8), 772–779. https://doi.org/10.1097/MLR.0000000000000560. Gruzd, A., Wellman, B., & Takhteyev, Y. (2011). Imagining Twitter as an imagined community. The American Behavioral Scientist, 55(10), 1294–1318. https://doi.org/ 10.1177/0002764211409378. Guo, C., & Saxton, G. D. (2014). Tweeting social change: How social media are changing nonprofit advocacy. Nonprofit and Voluntary Sector Quarterly, 43(1), 57–79. https://doi.org/10.1177/0899764012471585. Hauer, B. (2015). Why is everyone trying to #RaceTogether? [@iKadeeja]. March 15 [Tweet] Retrieved fromhttps://twitter.com/_iKadeeja. Hayes, T. J. (2017). MyNYPD: Transforming Twitter into a Public Place for Protest. Computers and Composition, 43, 118–134. Hayhurst, L. M. C., & Szto, C. (2016). Corporatizating activism through sport-focused social justice? Investigating Nike’s corporate responsibility initiatives in sport for development and peace. Journal of Sport & Social Issues, 40(6), 522–544. https://doi.org/10.1177/0193723516655579. Hazelton, V. (1992). Toward a systems theory of public relations. In H. Avenarius, & W. Armbrecht (Eds.). Is public relations a science? (pp. 33–55). Berlin: Westdeutscher Verlag. Heath, R. L., & Waymer, D. (2014). John Brown, public relations, terrorism, and social capital: “His Truth Goes Marching On”. Public Relations Inquiry, 3, 209–226. Hernandez, R. (2015). Here’s what people are saying about Starbucks’’ Race Together’ campaign. National Public Radiohttp://www.npr.org/sections/codeswitch/2015/ 03/17/393562838/heres-what-people-are-saying-about-starbucks-race-together-campaign. Hooker, J. (2016). Black lives matter and the paradoxes of U.S. Black politics: From democratic sacrifice to democratic repair. Political Theory, 44(4), 448–469. https:// doi.org/10.1177/0090591716640314. Isa, S. H. (2015). Not sure @Starbucks gets how they make racial tensions and propels the needs for BLM, #RaceTogether won’t fix it. [@beingbrina]. March 17 [Tweet]. Retrieved fromhttps://twitter.com/beingbrina. Jackson, S. J., & Foucault Welles, B. (2015). Hijacking #myNYPD: Social media dissent and networked counterpublics: Hijacking #myNYPD. The Journal of Communication, 65(6), 932–952. Jackson, L. A., von Eye, A., Fitzgerald, H. E., Zhao, Y., & Witt, E. A. (2010). Self-concept, self-esteem, gender, race and information technology use. Computers in Human Behavior, 26(3), 323–328. https://doi.org/10.1016/j.chb.2009.11.001. Johnston, J. (2015). Loose tweets sink fleets’ and other sage advice: Social media governance, policies and guidelines. Journal of Public Affairs, 15(2), 175–187. https:// doi.org/10.1002/pa.1538. Kanski, A. (2016). Starbucks: Positive newspaper ad blitz about much more than 2016 election. March 25, Retrieved fromPR Weekhttps://www.prweek.com/article/ 1389019/starbucks-positive-newspaper-ad-blitz-2016-election. Krage (2015). “@coreydu so are you unblocking/blocking people who are actually engaging you on a conversation abt race? How #racetogether is a stupid idea. [@Krageinsf]. March 20 [Tweet] Retrieved fromhttps://twitter.com/krageinsf?protected_redirect=true. LaMonica, P. (2015). Sorry Starbucks haters, #RaceTogether hasn’t hurt it. April 24, Retrieved fromCNN Moneyhttp://money.cnn.com/2015/04/24/investing/ starbucks-record-earnings-howard-schultz/index.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fmoney_latest+ %28CNNMoney%3A+Latest+News%29. Larson, E. D. (2016). Black lives matter and bridge building: Labor education for a “New Jim Crow” era. Labor Studies Journal, 41(1), 36–66. https://doi.org/10.1177/ 0160449X16638800. Liu, W., Lai, C., & Xu, W. (2018). Tweeting about emergency: A semantic network analysis of government organizations’ social media messaging during Hurricane Harvey. Public Relations Review, 44(5), 807–819. https://doi.org/10.1016/j.pubrev.2018.10.009. Logan, N. (2016). The Starbucks race together initiative: Analyzing a public relations campaign with critical race theory. Public Relations Inquiry, 5(1), 93–113. https:// doi.org/10.1177/2046147X15626969. Marwick, A. E., & boyd, D. (2011). I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society, 13(1), 114–133. https://doi.org/10.1177/1461444810365313. Men, L. R., Tsai, W.-H. S., Chen, Z. F., & Ji, Y. G. (2018). Social presence and digital dialogic communication: Engagement lessons from top social CEOs. Journal of Public Relations Research, 30(3), 83–99. Mengibar, A. C. (2015). Critical discourse analysis in the study of representation, identity politics and power relations: A multi-method approach. Comunicación y Sociedad, 28(2), 39. Minhaj, H. (2015). “Before they write #RaceTogether on cups, can Starbucks just spell my name correctly? [@Hasanminhaj]. March 17 [Tweet] Retrieved fromhttps:// twitter.com/hasanminhaj. Neuendorf, K. A., Skalski, P. D., Jeffres, L. W., & Atkin, D. (2014). Senses of humor, media use, and opinions about the treatment of marginalized groups. International Journal of Intercultural Relations, 42, 65–76. Nocera, J. (2015). Lessons from #RaceTogether. New York Timeshttps://www.nytimes.com/2015/04/21/opinion/joe-nocera-lessons-from-racetogether.html?_r=0. Novak, A., & El-Burki, I. J. (2016). Defining identity and the changing scope of culture in the digital age. Hershey, PA: IGI Global. Novak, A. N., Johnson, K., & Pontes, M. (2016). LatinoTwitter: Discourses of Latino civic engagement in social media. First Monday, 21(8). O’Callaghan, D., Greene, D., Conway, M., Carthy, J., & Cunningham, P. (2015). Down the (white) rabbit hole: The extreme right and online recommender systems. Social Science Computer Review, 33(4), 459–478. https://doi.org/10.1177/0894439314555329. Oluo, I. (2015). Being a barista is hard enough. Having to talk #RaceTogether with a woman in Lululemon pants while pouring pumpkin spice is just cruel. [@IjeomaOluo]. March 19 [Tweet] Retrieved fromhttps://twitter.com/IjeomaOluo. Paskevicius, M., Veletsianos, G., & Kimmons, R. (2018). Content is king: An analysis of how the twitter discourse surrounding open education unfolded from 2009 to 2016. International Review of Research in Open and Distance Learning, 19(1), 1–19. Peterson, H. (2015a). A Starbucks exec deleted his Twitter account after backlash over the company’s’ race together’ campaign. Business Insiderhttp://www.businessinsider. com/starbucks-race-together-campaign-2015-3. Peterson, H. (2015b). SNL skit hilariously mocks Starbucks’’ Race Together’ campaign. Business Insiderhttp://www.businessinsider.com/starbucks-snl-skit-2015-3. Peterson, H. (2015c). The real story behind Starbucks’ most embarrassing moment in history. Business Insider. http://www.businessinsider.com/starbucks-race-together- campaign-history-2015-6. Porter, M. C., Anderson, B., & Nhotsavang, M. (2015). Anti-social media: Executive twitter “engagement” and attitudes about media credibility. Journal of Communication Management, 19(3), 270–287. https://doi.org/10.1108/JCOM-07-2014-0041. Potts, T. J. (2012). Dark tourism’ and the ‘kitschification’ of 9/11. Tourist Studies, 12(3), 232–249. https://doi.org/10.1177/1468797612461083. Rasheed, A. (2015). What I want to do: 1. Eat Pizza 2. Pay off my loan debt… 77. Hang out with Rihanna… 895. Talk about race at Starbucks #RaceTogether.

10 A.N. Novak and J.C. Richmond Public Relations Review 45 (2019) 101773

[@AmmenaRasheed] March [Tweet] Retrieved fromhttps://twitter.com/ameenarasheed?lang=en. Ray, R., Brown, M., & Laybourn, W. (2017). The evolution of #BlackLivesMatter on twitter: Social movements, big data, and race. Ethnic and Racial Studies, 40(11), 1795–1796. https://doi.org/10.1080/01419870.2017.1335423. Rickford, R. (2016). Black lives matter: Toward a modern practice of mass struggle. New Labor Forum, 25(1), 34–42. https://doi.org/10.1177/1095796015620171. Sanders, S. (2015). Starbucks will stop putting the words’ Race Together’ on cups. National Public Radiohttp://www.npr.org/sections/thetwo-way/2015/03/22/ 394710277/starbucks-will-stop-writing-race-together-on-coffee-cups. Shah, K. (2015). Why Starbucks’ race together campaign failed. Eaterhttp://www.eater.com/2015/6/18/8807849/why-starbucks-race-together-campaign-failed. Sharma, S. (2013). Black twitter? Racial hashtags, networks and contagion. New Formations, 78, 46. https://doi.org/10.3898/NEWF.78.02.2013. Slush, J. (2015). As a stand up comedian, #RaceTogether is giving me the best material of my life. [@joshSlush] March 20 [Tweet] Retrieved fromThnx @Starbuckshttps:// twitter.com/JonSlush?lang=en. Some Place Simple (2015). If they [African Americans- derived from an earlier tweet] spent half as much time working as they did complaining on Twitter= no #RaceTogether. [@levijkm] March 21 [Tweet]. Retrieved fromhttps://twitter.com/levijkm. Stafford, Z. (2015). ’Barista: Your total is $5.45 me: You can just put that on my reparations tab. [@ZachStafford] March 17 [Tweet] Retrieved fromhttps://twitter.com/ zachstafford. Starbucks (2015). What ‘Race Together’ means for Starbucks partners and customers. Retrieved fromhttps://stories.starbucks.com/stories/2015/what-race-together- means-for-starbucks-partners-and-customers/. Starbucks Company Information (2019). Company information. Retrieved fromhttps://www.starbucks.com/about-us/company-information. Taranto, J. (2015). A nation of Howards; Starbucks’ bad idea provokes worse criticism. Wall Street Journal Online article: retrieved from https://www.wsj.com/ articles/a-nation-of-howards-1426795604 on 4/23/2019. Taylor, M., & Kent, M. L. (2014). Dialogic engagement: Clarifying foundational concepts. Journal of Public Relations Research, 26(5), 384–398. Thompson, L., & Cupples, J. (2008). Seen and not heard? Text messaging and digital sociality. Social & Cultural Geography, 9(1), 95–108. https://doi.org/10.1080/ 14649360701789634. Tracy the Clayton (2015). “Mocha money, mocha problems” “No chai left behind” “Black coffees matter” and “By any beans necessary. [@brokeymcpoverty] March 17 [Tweet] Retrieved fromhttps://twitter.com/brokeymcpoverty. Tynes, B. M., & Mitchell, K. J. (2014). Black youth beyond the digital divide: Age and gender differences in internet use, communication patterns, and victimization experiences. The Journal of Black Psychology, 40(3), 291–307. https://doi.org/10.1177/0095798413487555. Vidrine, D. J., Fletcher, F. E., Danysh, H. E., Marani, S., Vidrine, J. I., Cantor, S. B., et al. (2012). A randomized controlled trial to assess the efficacy of an interactive mobile messaging intervention for underserved smokers: Project ACTION. BMC Public Health, 12(1), https://doi.org/10.1186/1471-2458-12-696 696-696. Vos, T. P., & Li, Y. (2013). Justifying commercialization: Legitimating discourses and the rise of American advertising. Journalism & Mass Communication Quarterly, 90(3), 559–580. https://doi.org/10.1177/1077699013493787. Watkins, B. A. (2017). Experimenting with dialogue on Twitter: An examination of the influence of the dialogic principles on engagement, interaction, and attitude. Public Relations Review, 43(1), 163–171. Waymer, D. (Ed.). (2012). Culture, social class, and race in public relations: Perspectives and applications. Lanham, MD: Lexington Books. Werder, K. P., & Holtzhausen, D. (2009). An analysis of the influence of public relations department leadership style on public relations strategy use and effectiveness. Journal of Public Relations Research, 21(4), 404–427. https://doi.org/10.1080/10627260902966391. Whiting, A., & Williams, D. (2013). Why people use social media: A uses and gratifications approach. Qualitative Market Research an International Journal, 16(4), 362–369. Whitt, A. (2015). I think its good @Starbucks wants us to talk about #RaceTogether. Why is everyone upset that coffee is motivating us? [@AmandasWhitt] March 19 [Tweet] Retrieved fromhttps://twitter.com/amandaswhitt?lang=en. Williams, S. (2015). Digital defense: Black feminists resist violence with hashtag activism. Feminist Media Studies, 15(2), 341–344. Winter, N. J. G. (2014). Dangerous frames: How ideas about race and gender shape public opinion. Chicago: University of Chicago Press.